\begin{table}[!h]

\caption{Linear models on Republican violence (logged)\label{tab:lm_repub}}
\centering
\begin{threeparttable}
\begin{tabular}[t]{lccc}
\toprule
\multicolumn{1}{c}{ } & \multicolumn{1}{c}{Full sample} & \multicolumn{2}{c}{Reduced sample} \\
\cmidrule(l{3pt}r{3pt}){2-2} \cmidrule(l{3pt}r{3pt}){3-4}
\multicolumn{1}{c}{ } & \multicolumn{2}{c}{Clerics} & \multicolumn{1}{c}{All victims} \\
\cmidrule(l{3pt}r{3pt}){2-3} \cmidrule(l{3pt}r{3pt}){4-4}
  & (1) & (2) & (3)\\
\midrule
(Intercept) & -1.337*** & -1.831*** & -6.415***\\
 & (0.150) & (0.215) & (0.612)\\
Employers' association & 0.266*** & 0.435*** & 0.198+\\
 & (0.057) & (0.103) & (0.116)\\
CNT union & 0.066 & 0.022 & 0.183+\\
 & (0.044) & (0.067) & (0.106)\\
UGT union & 0.032 & 0.168* & 0.136*\\
 & (0.025) & (0.073) & (0.053)\\
\midrule
$n$ & 3948 & 1634 & 1634\\
$R^2$ & 0.44 & 0.46 & 0.56\\
Adj. $R^2$ & 0.43 & 0.45 & 0.55\\
Province FE & Yes & Yes & Yes\\
\bottomrule
\end{tabular}
\begin{tablenotes}[para]
\item \textit{Note: } 
\item + p $<$ 0.1, * p $<$ 0.05, ** p $<$ 0.01, *** p $<$ 0.001. Province FE not shown, clustered SE (province). Only municipalities that were at least one day under Republican control. Control variables not shown: population (log), leftist support 1936, days of republican control (log), illiteracy 1930, number of clerics (log). Model 2 is the same as Model 1 but restricts the sample to provinces where there is data available on Republican violence (Badajoz, Barcelona, Caceres, Ciudad Real, Girona, Huesca, Lleida, Asturias, Tarragona, Teruel, Zaragoza).
\end{tablenotes}
\end{threeparttable}
\end{table}
